Learning Graph Attentions via Replicator Dynamics.

Graph Attention (GA) which aims to learn the attention coefficients for graph edges has achieved impressive performance in GNNs on many graph learning tasks. However, existing GAs are usually learned based on edges' (or connected nodes') features which fail to fully capture the rich structural information of edges. Some recent research attempts to incorporate the structural information into GA learning but how to fully exploit them in GA learning is still a challenging problem. To address this challenge, in this work, we propose to leverage a new Replicator Dynamics model for graph attention learning, termed Graph Replicator Attention (GRA). The core of GRA is our derivation of replicator dynamics based sparse attention diffusion which can explicitly learn context-aware and sparse preserved graph attentions via a simple self-supervised way. Moreover, GRA can be theoretically explained from an energy minimization model. This provides a more theoretical justification for the proposed GRA method. Experiments on several graph learning tasks demonstrate the effectiveness and advantages of the proposed GRA method on ten benchmark datasets.

Novel Sinusoidal Signal Assisted Multivariate Variational Mode Decomposition Combined With Task-Related Component Analysis for Enhancing SSVEP-Based BCI Performance.

Brain-computer interfaces (BCIs) based on steady-state visually evoked potential (SSVEP) have a broad application prospect owing to their multiple command output and high performance. Each harmonic component of SSVEP individually contains unique features, which can be utilized to enhance the recognition performance of SSVEP-based BCIs. However, the existing subband analysis methods for SSVEP, including those based on filter banks and existing mode decomposition methods, have limitations in extracting and utilizing independent harmonic components. This study proposes a sinusoidal signal assisted multivariate variational mode decomposition (SA-MVMD) algorithm that allows the constraint of the center frequencies and narrowband filtering structures of the intrinsic mode functions (IMFs) based on the prior frequency knowledge of the signal. It preserves the target information of the signal during decomposition while avoiding mode mixing and incorrect decomposition, thereby enabling the effective extraction of each independent harmonic component of SSVEP. Building on this, a SA-MVMD based task-related component analysis (SA-MVMD-TRCA) method is further proposed to fully utilize the features within the overall SSVEP as well as its independent harmonics, thereby enhancing the recognition performance. Testing on the public SSVEP Benchmark dataset demonstrates that the proposed method significantly outperforms the filter bank-based control methods. This study confirms the effectiveness of SA-MVMD and the potential of this approach, which analyzes and utilizes each independent harmonic of SSVEP, providing new strategies and perspectives for performance enhancement in SSVEP-based BCIs.

An Eye Movement Classification Method based on Cascade Forest.

Eye tracking technology has become increasingly important in scientific research and practical applications. In the field of eye tracking research, analysis of eye movement data is crucial, particularly for classifying raw eye movement data into eye movement events. Current classification methods exhibit considerable variation in adaptability across different participants, and it is necessary to address the issues of class imbalance and data scarcity in eye movement classification. In the current study, we introduce a novel eye movement classification method based on cascade forest (EMCCF), which comprises two modules: (1) a feature extraction module that employs a multi-scale time window method to extract features from raw eye movement data; (2) a classification module that innovatively employs a layered ensemble architecture, integrating the cascade forest structure with ensemble learning principles, specifically for eye movement classification. Consequently, EMCCF not only enhanced the accuracy and efficiency of eye movement classification but also represents an advancement in applying ensemble learning techniques within this domain. Furthermore, experimental results indicated that EMCCF outperformed existing deep learning-based classification models in several metrics and demonstrated robust performance across different datasets and participants.

Clinical significance of preoperative nutritional status in elderly gastric cancer patients undergoing radical gastrectomy: A single-center retrospective study.

BACKGROUND: The population of elderly patients with gastric cancer is increasing, which is a major public health issue in China. Malnutrition is one of the greatest risk factors for adverse clinical outcomes in elderly patients with gastric cancer. AIM: To investigate the preoperative nutritional status and its association with delayed discharge of elderly gastric cancer patients following radical gastrectomy. METHODS: A total of 783 patients aged 65 years and older harboring gastric adenocarcinoma and following radical gastrectomy were retrospectively analyzed from the prospectively collected database of Zhongshan Hospital of Fudan University between January 2018 and May 2020. RESULTS: The overall rate of malnutrition was 31.8%. The incidence of postoperative complications was significantly higher in the malnourished group compared to the well-nourished group (P < 0.001). Nutritional characteristics in the malnourished group, including body mass index, prognostic nutritional index (PNI), albumin, prealbumin, and hemoglobin, were all significantly lower than those in the well-nourished group. The percentage of patients who received postoperative total nutrient admixture was lower in the malnourished group compared to the well-nourished group (22.1% vs 33.5%, P = 0.001). Age ≥ 70 years (HR = 1.216, 95%CI: 1.048-1.411), PNI < 44.5 (HR = 1.792, 95%CI: 1.058-3.032), operation time ≥ 160 minutes (HR = 1.431, 95%CI: 1.237-1.656), and postoperative complications grade III or higher (HR = 2.191, 95%CI: 1.604-2.991) were all recognized as independent risk factors associated with delayed discharge. CONCLUSION: Malnutrition is relatively common in elderly patients undergoing gastrectomy. Low PNI is an independent risk factor associated with delay discharge. More strategies are needed to improve the clinical outcome of these patients.

Insula-medial prefrontal cortex functional connectivity modulated by transcutaneous auricular vagus nerve stimulation: An fMRI study.

Transcutaneous auricular vagus nerve stimulation (taVNS) is an emerging neuro modulation technology that has been reported to be beneficial in the treatment of diseases by several studies, but its exact mechanism of action is still unclear. It has been demonstrated that ta VNS can influence interoceptive signals. Notably, the processing of interoceptive signals is directly related to many diseases, such as depression, anxiety, and insomnia. The insula and the medial prefrontal cortex (MPFC) communicate during the bottom-up transmission of taVNS-induced signals, and both play a role in interoceptive signal processing. By focusing on the insula and MPFC, our research pioneers detail the potential interactions between interoceptive signal processing and the neuromodulation effects of taVNS, providing novel insights in to the neurobiological mechanisms of taVNS. Two functional connectivity (FC) analyses (region of interest-based and seed-based) were used in this study. We observed that negative connectivity between the insula and the MPFC was significantly weakened following taVNS, while there were no statistical changes in the sham group. Our findings elucidate potential mechanisms linking vagal activity with intrinsic FC among specific brain regions and networks. Specifically, our results indicate that taVNS may enhance the ability to flexibly balance interoceptive awareness and cognitive experiences by modulating the FC between the insula and MPFC. The modulation effects may impact body-brain interactions, suggesting the mechanism of taVNS in therapeutic applications.

Within-Session Reliability of fNIRS in Robot-Assisted Upper-Limb Training.

Functional near-infrared spectroscopy (fNIRS) seems opportune for neurofeedback in robot-assisted rehabilitation training due to its noninvasive, less physical restriction, and no electromagnetic disturbance. Previous research has proved the cross-session reliability of fNIRS responses to non-motor tasks (e.g., visual stimuli) and fine-motor tasks (e.g., finger tapping). However, it is still unknown whether fNIRS responses remain reliable 1) in gross-motor tasks, 2) within a training session, and 3) for different training parameters. Hence, this study aimed to investigate the within-session reliability of fNIRS responses to gross-motor tasks for different training parameters. Ten healthy participants were recruited to conduct right elbow extension-flexion in three robot-assisted modes. The Passive mode was fully motor-actuated, while Active1 and Active2 modes involved active engagement with different resistance levels. FNIRS data of three identical runs were used to assess the within-session reliability in terms of the map- ( R2 ) and cluster-wise ( Roverlap ) spatial reproducibility and the intraclass correlation (ICC) of temporal features. The results revealed good spatial reliability ( R2 up to 0.69, Roverlap up to 0.68) at the subject level. Besides, the within-session temporal reliabilities of Slope, Max/Min, and Mean were between good and excellent ( ICC < 0.86). We also found that the within-session reliability was positively correlated with the intensity of the training mode, except for the temporal reliability of HbO in Active2 mode. Overall, our results demonstrated good within-session reliability of fNIRS responses, suggesting fNIRS as reliable neurofeedback for constructing closed-loop robot-assisted rehabilitation systems.

Transcatheter closure of perimembranous ventricular septal defects in children using a Wire-Drifting Technique

OBJECTIVE: Explore the feasibility and safety of transcatheter closure of perimembranous ventricular septal defects using a wire-drifting technique (WT) in children. METHODS: We retrospectively analyzed 121 pediatric patients diagnosed with perimembranous ventricular septal defects who underwent interventional treatment at the First Affiliated Hospital of Xi'an Jiaotong University from Dec 2011 to Dec 2014. Based on the method used for arteriovenous loop establishment during the procedure, the patients were divided into a conventional technique (CT) group and a WT group. RESULTS: In total, 51 of the 53 patients (96.2%) in the CT group and 66 of the 68 patients (97.1%) in the WT group achieved procedural success, with no significant difference between the two groups (p>0.05). The CT group showed a nonsignificantly higher one-time success rate of arteriovenous loop establishment (94.3% vs. 91.2%, p>0.05). The procedure time was 46.0 (14.0) min and 46.5 (10.0) min in the CT and WT groups, respectively. The CT procedure was discontinued in the 2 cases (3.8%) of intraprocedural atrioventricular block in the CT group. In the one case (1.9%) of postprocedural atrioventricular block in the CT group, a permanent pacemaker was implanted to resolve third-degree atrioventricular block three months after the procedure. In the WT group, no cases of intraprocedural atrioventricular block occurred, and one case (1.5%) of postprocedural atrioventricular block occurred. In this case, intravenous dexamethasone injection for three days returned the sinus rhythm to normal. Aggravated mild to moderate tricuspid regurgitation was observed in 2 patients (3.8%) in the CT group during the 2-year follow-up period; aggravated tricuspid regurgitation was not observed in the WT group. During the 2-year follow-up period, there was no evidence of residual shunting in either group. CONCLUSION: Transcatheter closure of perimembranous ventricular septal defects with the WT is safe and effective in children.